Paediatrica Indonesiana VOLUME 55
March
NUMBER 2
Original Article
Does malnutrition influence outcome in children undergoing congenital heart surgery in a developing country? Eva M. Marwali1, Sekarpramita Darmaputri1, Dadang H. Somasetia2, Sudigdo Sastroasmoro3, Nikolaus A. Haas4, Michael A. Portman5
Abstract
Background Most children undergoing cardiac surgery for
congenital heart disease (CHD) in developing countries are malnourished. Malnutrition is known as a co-morbidity factor that might predict and influence outcomes after surgery. Objectives To evaluate the effect of malnutrition and other associated risk factors on post-operative outcomes in children with CHDs underwent cardiac surgery. Methods We conducted a retrospective cohort study in a single center tertiary pediatric cardiac intensive care unit (PCICU) in Indonesia. Our cohort included all children between 5 and 36 months of age undergoing congenital heart surgery with cardiopulmonary bypass from November 2011 until February 2014. Outcomes measured were the length of intubation and the length of ICU stay. Variables for potential influence investigated were the nutritional status, age, gender, type of cardiac anomaly (acyanotic vs. cyanotic), Aristotle score, cardiopulmonary bypass time, aortic cross-clamp time, and Pediatric Risk of Mortality (PRISM) III score. Results Out of 249 patients included, 147 (59%) showed malnourishment on admission. Malnourished patients were significantly younger in age, presented with an acyanotic heart defects, and had higher PRISM III score. Additionally, they also had a longer mechanical ventilation time and ICU stay than those with a normal nutritional status. After adjusting for various variables using a multiple logistic regression model it could be demonstrated that a higher Z-score for weight to age was a significant protective factor for the intubation time of more than 29 hours with an odds ratio of 0.66 (95% CI 0.48 to 0.92, P = 0.012). Non-malnourished patients had a 49% significantly higher chance for extubation with a hazard ratio of 1.49 (95% CI 1.12 to 1.99, P= 0.007).
Conclusion Malnourishment is clearly associated in a linear
fashion with longer mechanical ventilation and ICU stay. As one of significant and potentially treatable co-morbidity factors, prevention of malnourishment by early diagnosis and optimal timing for surgery is important. [Paediatr Indones. 2015;55:109-16]. Keywords: malnutrition; congenital heart disease; cardiopulmonary bypass; intubation time
C
ongenital heart diseases (CHDs) are among the most frequent congenital anomalies presented at birth, with a reported incidence of 8 to 11/1,000 live
From the Division of Pediatric Cardiac ICU, National Cardiovascular Center Harapan Kita, Jakarta, Indonesia1, Division of Pediatric Emergency and Intensive Care, Department of Pediatrics, University of Padjadjaran/Hasan Sadikin Hospital, Bandung, Indonesia2, Center for Clinical Epidemiology and Evidence-based Medicine, Medical School University of Indonesia/Cipto Mangunkusumo Hospital, Jakarta, Indonesia3, Department of Congenital Heart Defect, Heart and Diabetes Center Northrhein Westfalia, Bad Oeynhausen, Germany4, Division of Cardiology, Seattle Children’s Hospital and Research Institute, University of Washington, Seattle, USA5. Reprint request to: Eva M. Marwali. Division of Pediatric Cardiac ICU, National Cardiovascular Center Harapan Kita, Jl. Letjen S. Parman Kav. 87 Slipi Jakarta 11420. Tel.+6221.568.4086. Fax. +6221.568.4230. E-mail:
[email protected].
Paediatr Indones, Vol. 55, No. 2, March 2015109
Eva M. Marwali et al: Malnutrition and outcome of congenital heart surgery
births, and with a high impact on neonatal morbidity and mortality (up to 33%).1 Additionally, several studies have consistently reported that malnutrition, ranging from mild undernutrition to complete failure to thrive,2 is a common cause of morbidity in children with CHD.3-5 Multiple factors contribute to growth impairment and malnutrition in infants and children with CHD, such as prenatal and genetic factors, hypoxia and hemodynamic factors, such as congestive heart failure, inadequate nutritional intake, swallowing dysfunction, gastroesophageal reflux, immaturity of the gastrointestinal tract, a hypermetabolic state, and nutrient malabsorption,6-8 as well as psychosocial and hormonal factors.9,10 However, in medically partially underserved nations such as Indonesia, poor access to care leads to late presentation of patients with CHD and also contributes to malnutrition. Children with CHD frequently require open heart surgery to repair the defect. However, to the authors’ knowledge, only limited studies have been conducted in developing countries focusing on the link between malnutrition and other factors on the post-surgical outcomes in children who undergo congenital heart surgery with cardiopulmonary bypass. The present study was performed to verify whether malnutrition was an independent risk factor for the length of mechanical ventilation and the length of intensive care unit (ICU) stay in this specific group of patients.
Methods This retrospective cohort study was performed from November 2011 until February 2014 and included all consecutive children aged 5 to 36 months admitted to the tertiary pediatric cardiac intensive care unit (PCICU) after congenital heart surgery with cardiopulmonary bypass. This age range was selected because this cohort showed homogenous characteristics and a tendency to require elective heart surgery over 5 months of age. We excluded patients with preexisting pulmonary disease, multiple congenital abnormalities, those who were subjected to tracheostomy, and those who died after the procedure. The nutritional status was determined by using
110Paediatr Indones, Vol. 55, No. 2, March 2015
the Z-score of weight for age (W/A), based on World Health Organization (WHO) child growth standard curves. 11 Malnourishment was defined in those patients in which W/A was at least two Z-scores below the mean standard value, otherwise patients were included in the study and grouped as normal. The nutritional status could not be determined by body height because no height data could be retrieved from our database. Other potential predictors for clinical outcomes, such as age, gender, the diagnosis of heart defect (acyanotic vs. cyanotic), Aristotle score (the score to indicate the complexity of surgery), cardiopulmonary bypass (CPB) time (the time when patient was on CPB machine), aortic cross-clamp time (the time when the aorta was clamped until the clamp was released), Pediatric Risk of Mortality (PRISM) III score (the score to predict mortality, consist of physical examination and laboratory results) on ICU admission were also assessed for their compromising effects with the nutritional status. The Glasgow coma scale was excluded from PRISM III score computation, as our patients were all sedated during ICU admission. The outcome variables were the length of mechanical ventilation and the length of ICU stay. As our patients varied in surgical complexities, the length of mechanical ventilation was categorized as prolonged if it was greater than 75th percentile of our cohort’s intubation time that was 29 hours post surgery. Power Analysis and Sample Size (PASS) for Windows software (version 11.0) was used to calculate the sample size needed to verify the association between malnutrition and length of intubation time using the Cox regression method; with 5% alpha error, 10% beta error, it yielded that a total of 193 patients were required, based on an expected 90% prevalence of malnutrition and expected relative morbidity risk of 1.70 in CHD children with a weight/ DJH=VFRUH11 All data collected were assessed for their normal distribution before being statistically analyzed. Subsequently, because the data distribution was not normal, we analyzed group differences using the Mann-Whitney test. We used x2 test and simple logistic regression to study the associations between the predictor variables (categorical and continuous variables, consecutively) and the binary outcome variables (length of intubation longer or shorter than 75th percentile). All significant predictor variables
Eva M. Marwali et al: Malnutrition and outcome of congenital heart surgery
were then analyzed in a multiple logistic regression model (forward stepwise). A Kaplan-Meier survival curve was generated to verify the influence of the nutritional status on the length of mechanical ventilation using the log rank test to compare the groups. Those which yielded P-value < 0.05 was considered as statistically significant. Data were expressed as mean and standard error of mean (SEM). All statistical analyses were performed using the SPSS for Mac software (version 21.0).
Results Out of a total of 1206 PCICU admissions (aged 0 to 18 years) from November 2011 until February 2014, only 547 patients were within the age range of 5-36 months. Of that number, 298 patients had incomplete or missing data. Hence, only 249 patients were eligible for the analysis presented. Our hospital had an overall 6.2% mortality rate for children aged 0 to 18 years undergoing congenital heart surgery. In our study,
Table 1. Main characteristics of patients studied Characteristics /GCPCIG 5'/ OQPVJU Male gender, n(%) /GCPDQF[YGKIJV 5'/ MI /GCP
